Changeset 11217 in ntrip


Ignore:
Timestamp:
Nov 20, 2026, 8:52:28 PM (less than one hour ago)
Author:
stuerze
Message:

Upload Ephemeris: majority vote improcved

Location:
trunk/BNC
Files:
3 edited

Legend:

Unmodified
Added
Removed
  • trunk/BNC/CHANGELOG.md

    r11216 r11217  
    22## 2.13.9 (unreleased)
    33- CHANGED: **Ephemerides of satellites outside the range supported by BNC** (e.g. G39, test signal of NTS-3 since 30 Sep 2026, not IS-GPS compliant, seen 19 Nov 2026 by many stations in Australia and the Pacific) are no longer written to RINEX navigation files or the ephemeris port, uploaded, or used in PPP, combination and SSR QC; each data set is logged once with its complete RINEX navigation record. Before, they were passed on unchanged (e.g. into an uploaded Broadcast Ephemeris stream), only the warning 'outside the range supported by BNC' was logged. RINEX Edit/Concatenate keeps such records of input files. New test case eph_out_of_range, test tool bncraw.py ephsats
    4 - ADDED: **Upload Ephemeris: majority vote over the copies of a data set.** A new data set (TOC and IOD) is uploaded as soon as three streams deliver the same content, at the latest 40 s after its first copy (before: the first copy immediately; SSR providers use a new IOD after 60 s, the 20 s left cover the upload interval and latency); meanwhile the previous data set stays in the upload (checked with each ephemeris and every second). Receivers and applications, BNC too, keep the first copy of a data set, so only this delay protects them; immediately uploaded if the previous data set came from less than three streams or for the first data set of a satellite. Replay 18 Nov 2026 (89 streams, 20 min): three agreeing streams after 14 s in the median, 87 % of 94 data sets within 40 s. Every further stream counts as a vote for its copy, and a copy with different content (the RTCM3 message without the GLONASS frame time tk and the system time offsets tau_c/tau_GPS, which receivers take from the string 5 of any satellite) replaces the uploaded one as soon as more streams deliver it than the current one, i.e. with at least three streams. Each replacement is logged with the streams per copy and the differing fields. Before, the fastest stream decided all fields of a data set, e.g. BCEP00BKG0 18 Nov 2026 R19/R20 with Bn = 1 (unhealthy) from RIO200ARG0/ENAO00PRT0 while all other streams sent Bn = 0, and ln3 = 1 from CUT000AUS0; replay of 89 input streams (2 h): GLONASS health different from the majority of the streams in 1 of 156 data sets (a 2:2 tie in Cn of the unhealthy R27) instead of 63. New test case upload_eph_vote
     4- ADDED: **Upload Ephemeris: majority vote over the copies of a data set.** A new data set (TOC and IOD) is uploaded as soon as three streams deliver the same content, at the latest 40 s after its first copy (before: the first copy immediately; SSR providers use a new IOD after 60 s, the 20 s left cover the upload interval and latency); meanwhile the previous data set stays in the upload (checked with each ephemeris and every second). Receivers and applications, BNC too, keep the first copy of a data set, so only this delay protects them; immediately uploaded if the previous data set came from less than three streams or for the first data set of a satellite. Replay 18 Nov 2026 (89 streams, 20 min): three agreeing streams after 14 s in the median, 87 % of 94 data sets within 40 s. Every further stream counts as a vote for its copy, and a copy with different content (the RTCM3 message without the GLONASS frame time tk and the system time offsets tau_c/tau_GPS, which receivers take from the string 5 of any satellite) replaces the uploaded one as soon as more streams deliver it than the current one, i.e. with at least three streams (only the latest copy per stream counts, so the only stream of a data set may replace its own copy). Each replacement is logged with the streams per copy and the differing fields. Before, the fastest stream decided all fields of a data set, e.g. BCEP00BKG0 18 Nov 2026 R19/R20 with Bn = 1 (unhealthy) from RIO200ARG0/ENAO00PRT0 while all other streams sent Bn = 0, and ln3 = 1 from CUT000AUS0; replay of 89 input streams (2 h): GLONASS health different from the majority of the streams in 1 of 156 data sets (a 2:2 tie in Cn of the unhealthy R27) instead of 63. New test case upload_eph_vote
    55- ADDED: **SSR QC: code biases of APC and CoM streams.** For IGS-SSR and old RTCM-SSR (not new RTCM-SSR, which carries its own satellite antenna model), the code biases of an APC stream (SSRA...) are compared with those of the CoM stream of the same solution and format (SSRC...); following the IGS SSR convention (users reduce the satellite PCOs of all frequencies by the PCO of the reference frequency) they have to be equal. With 'QC ANTEX File' the differences are compared with the PCO differences z(f) - z(ref): 15 Oct 2026 SSRA00GMV0 GPS/Galileo/BDS biases shifted by z(f) - z(ref) (residuals 2/3/14 mm, up to 0.76 m, B_IF of GPS IIIA 0.78 m), GLONASS equal. New test case ssr_qc_apc_com
    66- ADDED: **SSR QC: phase bias indicators and Melbourne-Wuebbena check.** Per stream and system the SSR format, the dispersive bias and MW consistency indicators, the integer and wide-lane integer indicators per signal and the satellites with discontinuity counter changes are reported. Every 300 s the MW combination of the phase and code biases of the reference signals (wide-lane cycles) is differenced with that of each other AC; the RMS of the fractional parts about the common offset shows whether the wide-lane biases are consistent (< 0.10 cycles) as required for a PPP-AR combination. 2 Oct 2026: MW consistency indicator 1 only for CNES and NRCan, CNES vs NRCan GPS 0.14 cycles (single satellites up to 0.3), all other pairs ~0.25 cycles (random). New test case ssr_qc_phase_bias
  • trunk/BNC/src/bncephuser.h

    r11216 r11217  
    5858  // buffered immediately; a copy with different content (RTCM3 encoding without
    5959  // the GLONASS frame time tk and the system time offsets tau_c, tau_GPS) replaces it as soon as more streams deliver it
    60   // than the current one, i.e. with at least three streams.
     60  // than the current one, i.e. with at least three streams (only the latest
     61  // copy of each stream counts: a single stream may replace its own copy).
    6162  // A new data set is held back from the upload (ephUpload() returns the
    6263  // previous one) until three streams deliver the same content or VOTE_HOLD
  • trunk/BNC/src/bnchelp.html

    r11216 r11217  
    1007210072    first copy received is buffered; every further stream counts as a vote for the copy it delivers, and
    1007310073    a copy with different content replaces the buffered one as soon as more streams deliver it than the current one,
    10074     i.e. with at least three streams. A new data set is uploaded as soon as three streams deliver the same content as
     10074    i.e. with at least three streams; only the latest copy of each stream counts, so a stream which is the only
     10075    source of a data set and changes its content replaces its own copy (seen for Galileo I/NAV BGDs of ENAO00PRT0,
     10076    20 Nov 2026). A new data set is uploaded as soon as three streams deliver the same content as
    1007510077    the buffered copy, at the latest 40 seconds after its first copy; until then the previous data set of the
    1007610078    satellite stays in the upload; the hold is checked with each ephemeris received and every second. Data sets of satellites whose previous data set came from less than three streams
Note: See TracChangeset for help on using the changeset viewer.